IT AI Index
Index › Data platform › September 2026 Edition

ML platforms

Asked as “machine learning platform”, and as “MLOps platform for training and deploying models”, on behalf of a mid-market B2B company. 48 first choices recorded across the direct, paraphrase, budget and scale prompts, twelve models each.
Standing
Azure Machine Learning
Contested
17% of first choices, contested.
ShowinginAll 67 categories →

By buyer segment

The same question asked on behalf of a different buyer. Each standing is computed within its segment; they sit side by side and are never added together.
Small businessGoogle Vertex AI17%ContestedStanding →Mid-marketAzure Machine Learning17%ContestedThis pageEnterpriseDatabricks26%ContestedStanding →

01The standing

Share is the count of first choices across the direct, paraphrase, budget and scale prompts, over all twelve models, for a mid-market B2B company. Ordered by share.
ProductFirst-choice shareNegative rateLabelsQuadrant
01Azure Machine Learning17%18%49accepted challenger
02Databricks12%10%30accepted challenger
03Amazon SageMaker10%29%49criticized challenger
04Google Vertex AI8%20%44accepted challenger
05Dataiku6%0%15accepted challenger
06Databricks Mosaic AI6%0%10accepted challenger
07TensorFlow4%0%10accepted challenger
08MLflow2%12%25accepted challenger
Show the four products at 0%, ordered by negative rate
12Kubeflow0%38%13criticized challenger
11Weights & Biases0%25%12criticized challenger
09DataRobot0%16%19accepted challenger
10H2O.ai0%13%15accepted challenger
Bars are the share of first choices, 0 to 100Every product with at least 10 labels here. Every product name links to its vendor page.
Head to head
Azure Machine Learning vs DatabricksAzure Machine Learning vs Amazon SageMakerDatabricks vs Amazon SageMaker

Recommended versus criticized

Every product with at least 10 labels here, on both axes. The 30% line names a quadrant, not the verdict above: that one needs more than 40%.

Criticized challengerCriticized default
Negative label rate →
01
02
03
04
05
06
07
08
09
10
11
12
Accepted challengerEndorsed leader
0%First-choice share → · lines at 30% share and 25% negative30%
Key
01Azure Machine Learning17%
02Databricks12%
03Amazon SageMaker10%
04Google Vertex AI8%
05Dataiku6%
06Databricks Mosaic AI6%
07TensorFlow4%
08MLflow2%
09DataRobot0%
10H2O.ai0%
11Weights & Biases0%
12Kubeflow0%

02What they warned about

Two of twelve models held their first choice under the paraphrase. Claude Haiku 4.5, Gemini 3.5 Flash, Perplexity Sonar, Grok 4.1 Fast, Mistral Small, DeepSeek V4 Flash, Qwen 3.7 Flash, Kimi K2, GLM 4.7 FlashX and MiniMax M2.5 changed. A high negative share on a product with few labels is a warning. A low share on a product with many labels is salience, not sentiment.
Amazon SageMaker
29%
14 of 49 labels negative · 10 of 12 models · 1 hard negative
“Avoid these platforms if you are a small team, a startup, or just starting out.” GLM 4.7 FlashX, negative prompt
Google Vertex AI
20%
9 of 44 labels negative · 8 of 12 models · 1 hard negative
“Avoid these platforms if you are a small team, a startup, or just starting out.” GLM 4.7 FlashX, negative prompt
Azure Machine Learning
18%
9 of 49 labels negative · 7 of 12 models · 1 hard negative
“Avoid these platforms if you are a small team, a startup, or just starting out.” GLM 4.7 FlashX, negative prompt
DeepSeek
100%
4 of 4 labels negative · 4 of 12 models · 4 hard negative
“leading experts to advise against logging in or using the service for enterprise tasks until stability improves” Qwen 3.7 Flash, negative prompt

03What they cite

Citations exist only for the models that return a source list: twelve of the twelve in this edition, and all six flagship models on the expanded tier.

Sites the answers cite

64 of 72 answers in this category came back with a source list, from 12 of 12 models: citations where the model returns them, or the search results it consulted. 6 of those lists are Google grounding redirects that name no site and are left out of the counts. 785 links across 293 sites, every framing counted. Ranked by the number of answers carrying the site or page.

g2.com
20 answers · 33 citations · 9 models
trustradius.com
20 answers · 28 citations · 10 models
learn.g2.com
17 answers · 23 citations · 10 models
techtarget.com
12 answers · 15 citations · 8 models
unite.ai
12 answers · 13 citations · 7 models
devopsschool.com
11 answers · 14 citations · 9 models
deepusecase.com
10 answers · 20 citations · 6 models
articsledge.com
10 answers · 11 citations · 6 models
dagshub.com
vendor site · DagsHub10 answers · 10 citations · 7 models
dataiku.com
vendor site · Dataiku10 answers · 10 citations · 8 models
techvendorindex.com
9 answers · 14 citations · 9 models
aiadvisorypractice.com
8 answers · 10 citations · 6 models

Pages the answers cite

The ten pages named in the most answers, by full address. A page here is one the models returned with a recommendation, not one the index endorses.

learn.g2.com/best-data-science-ml-platforms
14 answers · 9 models
unite.ai/10-best-machine-learning-software
12 answers · 7 models
dataiku.com/blog/enterprise-machine-learning-platforms-how-to-choose
10 answers · 8 models
dagshub.com/blog/top-deep-learning-platforms
9 answers · 7 models
techvendorindex.com/compare/best-ai-ml-for-mid-market
9 answers · 9 models
trustradius.com/machine-learning?company-size=mid-sized
9 answers · 9 models
articsledge.com/post/machine-learning-platforms-comparison-azure-ai-vs-aws-…
8 answers · 5 models
checkthat.ai/answers/what-are-the-best-machine-learning-platforms
8 answers · 5 models
learn.g2.com/best-machine-learning-tools
8 answers · 7 models
thirstysprout.com/post/best-machine-learning-platforms
8 answers · 7 models

04How they answered

Six framings of the same buying question, each sent to every model in a fresh session with search on. One row per model, so a row shows whether it held its answer under rewording, what it named when cost was the constraint, and what it argued against. Computed from the raw judge labels.
ModelDirect“What is the best machine learning platform for a mid-market B2B company?”Paraphrase“Which MLOps platform for training and deploying models would you recommend to a mid-sized B2B company?”Comparative“What are the top machine learning platforms and how do they differ?”Budget-constrained“What is the best machine learning platform for a company with a limited budget?”Scale-constrained“We are a 500 person company evaluating a machine learning platform. What should we look at?”Negative“Which machine learning platforms should I avoid or be cautious about?”
Claude Haiku 4.5no first choiceDatabricks Mosaic AIChanged
Four alternativesAzure Machine Learning, DataRobot, Google Vertex AI, TrueFoundry
no first choiceGoogle Cloud AutoML & Vertex AI, TensorFlow, scikit-learn
Four alternativesAmazon SageMaker, Azure Machine Learning, H2O AutoML, PyTorch
no first choiceagainst: Deep Lake, Lunary, NVIDIA's NeMo, Weights & Biases' Weave toolkit, ZenML
GPT-5.4 miniDatabricks
Two alternativesAmazon SageMaker, Azure Machine Learning
DatabricksHeld
One alternativeGoogle Vertex AI
against: Amazon SageMakerno first choice
Eight alternativesAmazon SageMaker, Azure Machine Learning, Databricks, Google Vertex AI, H2O.ai, IBM watsonx.ai, MLflow, Snowflake ML / Cortex AI
Azure Machine Learning
One alternativeAmazon SageMaker
against: Databricks, Snowflake MLno first choicenothing named
Gemini 3.5 FlashDataiku
Five alternativesAzure Machine Learning, DataRobot, Databricks, Pecan AI, Snowflake
against: Amazon SageMakerClearMLChanged
Five alternativesAmazon SageMaker, Databricks, Google Vertex AI, MLflow, TrueFoundry
against: Kubeflowno first choice
Nine alternativesAmazon SageMaker, Azure Machine Learning, DataRobot, Databricks, Dataiku, Google Vertex AI, MLflow, Snowflake, Weights & Biases
ClearML
Seven alternativesAmazon SageMaker, Google Vertex AI, Hugging Face, KNIME Analytics Platform, MLflow, Paperspace, Saturn Cloud
against: Alteryx, Amazon Neptune, DataRobot, Weights & BiasesDatabricks Mosaic AI
Five alternativesAmazon SageMaker, Braintrust, Google Vertex AI, LangSmith, Weights & Biases
against: Kubeflowagainst: Amazon SageMaker, Azure Machine Learning, DataRobot, Google Vertex AI, H2O Driverless AI, IBM Watson Studio
Perplexity SonarAzure Machine Learning
Three alternativesDatabricks Mosaic AI, Dataiku, Snowflake Cortex AI
Databricks Mosaic AIChanged
Four alternativesAmazon SageMaker, BentoML, Google Vertex AI, MLflow
no first choice
Six alternativesAmazon SageMaker, Azure Machine Learning, DataRobot, Databricks, Google Vertex AI, H2O.ai
Google Colab, KNIME
Three alternativesBigML, PyCaret, scikit-learn
no first choiceagainst: Amazon SageMaker, Azure Machine Learning, DagsHub, Google Vertex AI, H2O.ai
Grok 4.1 FastAzure Machine Learning
Three alternativesDataRobot, Databricks Mosaic AI, Dataiku
against: H2O.aiAmazon SageMaker, Azure Machine Learning, Google Vertex AIChanged
Two alternativesDatabricks, MLflow
no first choiceGoogle Colab, Kaggle
Three alternativesHugging Face, KNIME Analytics Platform, PyTorch/TensorFlow + local setup
against: Amazon SageMaker, Azure Machine Learning, Google Vertex AIno first choiceagainst: TensorFlow/PyTorchagainst: Amazon SageMaker, Azure Machine Learning, ChatGPT, ClearML, Copilot, DeepSeek, Google Gemini, Google Vertex AI, Hugging Face, MLflow, Meta AI
Mistral SmallAzure Machine Learning
Four alternativesAmazon SageMaker, Databricks Mosaic AI, Dataiku, Google Vertex AI
Amazon SageMaker, DatabricksChanged
Two alternativesGoogle Vertex AI, MLflow
no first choiceKNIME, Kaggle
Three alternativesAzure Machine Learning, BigML, Google Vertex AI
no first choiceagainst: Amazon SageMaker, Azure Machine Learning, DagsHub, DeepSeek, Google Vertex AI
DeepSeek V4 FlashAmazon SageMaker, Azure Machine Learning, Google Vertex AI
Four alternativesDataRobot, Dataiku, H2O.ai, KNIME
DatabricksChanged
Four alternativesAmazon SageMaker, Azure Machine Learning, Google Vertex AI, MLflow
against: KubeflowAmazon SageMaker
Eight alternativesAzure Machine Learning, DataRobot, Databricks, Google Vertex AI, PyTorch, RapidMiner, TensorFlow, scikit-learn
Google Vertex AI
Six alternativesAmazon SageMaker, Azure ML Designer, Google Colab, Google Vertex AutoML, MLflow, PyTorch
no first choiceagainst: Amazon SageMaker, Azure Machine Learning, C3 AI, DataRobot, Databricks, Google Vertex AI, MLflow, MindsDB, Palantir, TensorFlow Probability, Weights & Biases
Llama 4 Maverickno first choiceno first choiceHeldno first choiceKNIME, TensorFlow
Two alternativesBigML, Google Vertex AI
no first choiceagainst: Amazon SageMaker, DagsHub, Google Vertex AI, Kubeflow
Qwen 3.7 FlashAzure Machine Learning
Two alternativesAmazon SageMaker, Google Vertex AI
DatabricksChanged
Two alternativesAzure Machine Learning, Domino Data Lab
against: Amazon SageMakerno first choiceHugging Face
Four alternativesGoogle Colab, Kaggle, MLflow, Weights & Biases
against: Amazon SageMaker, Azure Machine Learningno first choiceagainst: Amazon SageMaker, Azure Machine Learning, DagsHub, DeepSeek, Google Colab, Google Vertex AI, Kubeflow
Kimi K2Dataiku
Two alternativesAzure Machine Learning, Databricks Mosaic AI
Amazon SageMakerChanged
Two alternativesAzure Machine Learning, Google Vertex AI
no first choice
Seven alternativesAmazon SageMaker, Azure Machine Learning, DataRobot, Google Vertex AI, PyTorch, TensorFlow, scikit-learn
MLflow
Five alternativesKNIME Analytics Platform, PyTorch, TensorFlow + Keras, Weights & Biases, scikit-learn
no first choiceagainst: Databricks, OpenAI, Weights & Biases' Weave toolkit, ZenML
GLM 4.7 FlashXDataiku
Three alternativesAzure Machine Learning, DataRobot, Databricks
Google Vertex AIChanged
Six alternativesAmazon SageMaker, Azure Machine Learning, ClearML, Kubeflow, MLflow, Weights & Biases
no first choice
Eight alternativesAmazon SageMaker, Azure Machine Learning, DataRobot, Databricks, Google Vertex AI, H2O.ai, Kubeflow, MLflow
Jupyter Notebook, KNIME Analytics Platform, Orange
Nine alternativesAzure Machine Learning, Databricks, Google Colab, Google Vertex AI, KNIME, Kaggle, PyTorch, TensorFlow, scikit-learn
against: Zoho Analyticsno first choiceagainst: Amazon SageMaker, Azure Machine Learning, Google Vertex AI, Hugging Face, MLflow, Weights & Biases
MiniMax M2.5Azure Machine Learning, Databricks
Two alternativesGoogle Vertex AI, Snowflake Cortex AI
against: Amazon SageMakerAmazon SageMakerChanged
Three alternativesDatabricks, Kubeflow, MLflow
no first choice
Six alternativesAmazon SageMaker, Azure Machine Learning, Databricks Mosaic AI, Dataiku, Google Vertex AI, H2O.ai
KNIME Analytics Platform
Three alternativesAzure Machine Learning, TensorFlow, Weka
no first choiceagainst: DeepSeek, Google Gemini, Meta AI, Microsoft Copilot
Bold is the first choiceAlternatives are counted; the count opens them.What the answer argued against

05The record

One row per call: the version string exactly as returned, whether the model searched, sources cited, and latency. Full answer text is in the free responses file. Download the record
Seventy-two rows: every prompt, every model, every answer.
PromptModelVersion stringTime (UTC)SearchedSourcesLatency
Direct recommendationClaude Haiku 4.5claude-haiku-4-5-202510012026-09-17 11:40no03 s
Direct recommendationGPT-5.4 minigpt-5.4-mini-2026-03-172026-09-17 09:53yes46 s
Direct recommendationGemini 3.5 Flashgemini-3.5-flash2026-09-17 10:13yes1526 s
Direct recommendationPerplexity Sonarsonar2026-09-17 12:59yes205 s
Direct recommendationGrok 4.1 Fastspacexai/grok-4.1-fast-non-reasoning via vertex2026-09-17 11:03yes3521 s
Direct recommendationMistral Smallmistral/mistral-small via mistral2026-09-17 13:23yes118 s
Direct recommendationDeepSeek V4 Flashdeepseek/deepseek-v4-flash via fireworks2026-09-17 10:48yes1921 s
Direct recommendationLlama 4 Maverickmeta/llama-4-maverick via bedrock2026-09-17 12:25yes53 s
Direct recommendationQwen 3.7 Flashalibaba/qwen3.7-flash via alibaba2026-09-17 10:56yes1056 s
Direct recommendationKimi K2moonshotai/kimi-k2 via novita2026-09-17 11:04yes934 s
Direct recommendationGLM 4.7 FlashXzai/glm-4.7-flashx via zai2026-09-17 10:03yes21262 s
Direct recommendationMiniMax M2.5minimax/minimax-m2.5 via minimax2026-09-17 09:59yes826 s
ParaphraseClaude Haiku 4.5claude-haiku-4-5-202510012026-09-17 11:44yes1812 s
ParaphraseGPT-5.4 minigpt-5.4-mini-2026-03-172026-09-17 12:10yes57 s
ParaphraseGemini 3.5 Flashgemini-3.5-flash2026-09-17 13:26yes1624 s
ParaphrasePerplexity Sonarsonar2026-09-17 13:12yes208 s
ParaphraseGrok 4.1 Fastspacexai/grok-4.1-fast-non-reasoning via vertex2026-09-17 11:57yes3120 s
ParaphraseMistral Smallmistral/mistral-small via mistral2026-09-17 11:19yes55 s
ParaphraseDeepSeek V4 Flashdeepseek/deepseek-v4-flash via fireworks2026-09-17 09:31yes1514 s
ParaphraseLlama 4 Maverickmeta/llama-4-maverick via bedrock2026-09-17 13:00yes54 s
ParaphraseQwen 3.7 Flashalibaba/qwen3.7-flash via alibaba2026-09-17 12:40no038 s
ParaphraseKimi K2moonshotai/kimi-k2 via novita2026-09-17 13:17yes1549 s
ParaphraseGLM 4.7 FlashXzai/glm-4.7-flashx via zai2026-09-17 12:06yes2138 s
ParaphraseMiniMax M2.5minimax/minimax-m2.5 via minimax2026-09-17 09:47yes1029 s
ComparativeClaude Haiku 4.5claude-haiku-4-5-202510012026-09-17 10:01yes99 s
ComparativeGPT-5.4 minigpt-5.4-mini-2026-03-172026-09-17 10:58yes510 s
ComparativeGemini 3.5 Flashgemini-3.5-flash2026-09-17 11:49yes1332 s
ComparativePerplexity Sonarsonar2026-09-17 10:37yes207 s
ComparativeGrok 4.1 Fastspacexai/grok-4.1-fast-non-reasoning via vertex2026-09-17 13:42yes2814 s
ComparativeMistral Smallmistral/mistral-small via mistral2026-09-17 10:16yes56 s
ComparativeDeepSeek V4 Flashdeepseek/deepseek-v4-flash via fireworks2026-09-17 11:13yes2019 s
ComparativeLlama 4 Maverickmeta/llama-4-maverick via bedrock2026-09-17 13:47yes52 s
ComparativeQwen 3.7 Flashalibaba/qwen3.7-flash via alibaba2026-09-17 09:56yes531 s
ComparativeKimi K2moonshotai/kimi-k2 via novita2026-09-17 11:54yes1029 s
ComparativeGLM 4.7 FlashXzai/glm-4.7-flashx via zai2026-09-17 12:11yes3574 s
ComparativeMiniMax M2.5minimax/minimax-m2.5 via minimax2026-09-17 11:00yes516 s
Budget constrainedClaude Haiku 4.5claude-haiku-4-5-202510012026-09-17 12:04no06 s
Budget constrainedGPT-5.4 minigpt-5.4-mini-2026-03-172026-09-17 11:28yes57 s
Budget constrainedGemini 3.5 Flashgemini-3.5-flash2026-09-17 10:51yes921 s
Budget constrainedPerplexity Sonarsonar2026-09-17 10:00yes205 s
Budget constrainedGrok 4.1 Fastspacexai/grok-4.1-fast-non-reasoning via vertex2026-09-17 14:00yes149 s
Budget constrainedMistral Smallmistral/mistral-small via mistral2026-09-17 11:03yes54 s
Budget constrainedDeepSeek V4 Flashdeepseek/deepseek-v4-flash via fireworks2026-09-17 12:55yes1920 s
Budget constrainedLlama 4 Maverickmeta/llama-4-maverick via bedrock2026-09-17 12:00yes53 s
Budget constrainedQwen 3.7 Flashalibaba/qwen3.7-flash via alibaba2026-09-17 13:46no034 s
Budget constrainedKimi K2moonshotai/kimi-k2 via novita2026-09-17 11:21yes1026 s
Budget constrainedGLM 4.7 FlashXzai/glm-4.7-flashx via zai2026-09-17 12:13yes1071 s
Budget constrainedMiniMax M2.5minimax/minimax-m2.5 via minimax2026-09-17 11:30yes525 s
Scale constrainedClaude Haiku 4.5claude-haiku-4-5-202510012026-09-17 10:07no06 s
Scale constrainedGPT-5.4 minigpt-5.4-mini-2026-03-172026-09-17 12:07no08 s
Scale constrainedGemini 3.5 Flashgemini-3.5-flash2026-09-17 13:10yes1230 s
Scale constrainedPerplexity Sonarsonar2026-09-17 12:24yes207 s
Scale constrainedGrok 4.1 Fastspacexai/grok-4.1-fast-non-reasoning via vertex2026-09-17 09:52yes128 s
Scale constrainedMistral Smallmistral/mistral-small via mistral2026-09-17 11:16yes58 s
Scale constrainedDeepSeek V4 Flashdeepseek/deepseek-v4-flash via fireworks2026-09-17 12:07yes1819 s
Scale constrainedLlama 4 Maverickmeta/llama-4-maverick via bedrock2026-09-17 13:54yes53 s
Scale constrainedQwen 3.7 Flashalibaba/qwen3.7-flash via alibaba2026-09-17 13:41no036 s
Scale constrainedKimi K2moonshotai/kimi-k2 via novita2026-09-17 11:19yes1323 s
Scale constrainedGLM 4.7 FlashXzai/glm-4.7-flashx via zai2026-09-17 09:29yes1274 s
Scale constrainedMiniMax M2.5minimax/minimax-m2.5 via minimax2026-09-17 14:01yes520 s
Negative framingClaude Haiku 4.5claude-haiku-4-5-202510012026-09-17 12:58yes188 s
Negative framingGPT-5.4 minigpt-5.4-mini-2026-03-172026-09-17 10:08no05 s
Negative framingGemini 3.5 Flashgemini-3.5-flash2026-09-17 13:36yes1925 s
Negative framingPerplexity Sonarsonar2026-09-17 11:38yes207 s
Negative framingGrok 4.1 Fastspacexai/grok-4.1-fast-non-reasoning via vertex2026-09-17 10:44yes4216 s
Negative framingMistral Smallmistral/mistral-small via mistral2026-09-17 09:23yes56 s
Negative framingDeepSeek V4 Flashdeepseek/deepseek-v4-flash via fireworks2026-09-17 11:43yes2020 s
Negative framingLlama 4 Maverickmeta/llama-4-maverick via bedrock2026-09-17 14:07yes54 s
Negative framingQwen 3.7 Flashalibaba/qwen3.7-flash via alibaba2026-09-17 10:25yes540 s
Negative framingKimi K2moonshotai/kimi-k2 via novita2026-09-17 13:22yes1524 s
Negative framingGLM 4.7 FlashXzai/glm-4.7-flashx via zai2026-09-17 11:26yes3377 s
Negative framingMiniMax M2.5minimax/minimax-m2.5 via minimax2026-09-17 12:46yes523 s

Normalization in this category

Every judgment call made between the raw labels and the numbers above, listed so it is visible and reversible.

Category-scoped readings
Domino read as Domino Data Lab
Unresolved, counted raw
6sense
AWS SageMaker Studio (Free Tier)
Amplemarket
Azure ML Designer
Deep Lake
Google Cloud AI/ML
Google Vertex AutoML
Grok
HubSpot AI
Le Chat
Lunary
MindsDB
NVIDIA's NeMo
PyTorch/TensorFlow + local setup
RStudio (Open Source)
Seldon Core
Snowflake ML / Cortex AI
TensorFlow + Keras
TensorFlow Probability
XGBoost / LightGBM / CatBoost
Zoho Analytics
ZoomInfo
Discontinued, still offered
No shut-down product was recommended here.

Also in data platform

The other categories of this function, with the leader in each.
← Managed databasesNoSQL →